[Collab-maint-devel] Re: Collaborative maintenance, time to work

[Otavio Salvador]

Daniel Ruoso <daniel at ruoso.com> writes:

> usually the untagged is probably broken... when something is finished,
> the changelog entry and the tag are created (I work this way, actually,
> I don't know if this is the common practice) I like to commit a lot even
> before the work is completely done...

More or less. I think everything that has a debian/changelog with a
UNRELEASED suite setted could be exported that way. So, if I don't
have to have it exported, I shouldn't have a changelog that make this
package buildable. I think this is enough for major of user cases.

>> Tagged versions are usually uploaded shortly, and thus this APT repository
>> wouldn't be more bleeding-edge than unstable/experimental ...
>
> except that it's possible to imagine that each group could have it's
> experimental sources available before it goes to unstable... I mean, a
> transition can be made with less burden on the debian archive, since the
> developers can have a source repository to mess with and test better
> before uploading to ftp-master, then uploading a full set of already
> tested packages...

But then wouldn't be better to upload to experimental? There we have
more users to test it. I think is good to have a "snapshot" repository
that's always in sync with development sources so their involved
people could use while it isn't ready for production use yet.

-- 
        O T A V I O    S A L V A D O R
---------------------------------------------
 E-mail: otavio at debian.org      UIN: 5906116
 GNU/Linux User: 239058     GPG ID: 49A5F855
 Home Page: http://www.freedom.ind.br/otavio
---------------------------------------------
"Microsoft gives you Windows ... Linux gives
 you the whole house."